REQUERIMIENTOS BÁSICOS PARA UNA PROTECCIÓN

The action of cardiac vagal preganglionic C-fibre efferents was not determined from central recordings (chapter two). In order to study the possible chronotropic action of C-fibre efferents, direct selective stimulation of this group is necessary.

Previous studies attempting a central electrical DVMN stimulation have either failed or afforded unconvincing evidence to support the hypothesis that this is a "cardioinhibitory centre". This approach is to be avoided for three reasons: first the axons of B-fibre cardiac vagal preganglionic neurones lie close to the axons of the C-fibre efferents, therefore electrical stimulation will yield a false positive result, second the C-fibre efferents are scattered along a considerable longitudinal stretch of neuraxis, therefore focal stimulation may yield a false negative result. Third, since the DVMN is so close to NTS neurites , current spread and electrical stimulation of NTS pathways involved in reflex cardioinhibition may account for a false positive result.

Early studies on the effect of vagal stimulation employed the use of induction coils and this was subsequently combined with recordings of evoked compound action potentials displayed on cathode ray oscilloscopes. To assess the contribution of unmyelinated preganglionic axons, a stimulation intensity was chosen which activated only myelinated axons and the effect on heart rate noted. By increasing the stimulus intensity the unmyelinated fibres were recruited and any additional effect on heart rate was then ascribed to their action on the ganglion. This approach therefore, aimed to compare the effects of B-fibres versus B+C fibres.

This approach is to be avoided for four reasons: First, increasing stimulus intensity increases B-fibre recruitment therefore B+C should be greater than B alone and this has nothing to do with C-fibre recruitment, (a false positive). Second, the stimulus intensity

which activates B-fibres may produce a near maximal response and the effect of C-fibre recruitment may be masked (a false negative). Third heart rate rather than cardiac cycle length (R-R interval) has been recorded in most vagal stimulation studies. It is possible to produce enormous R-R interval changes and minuscule heart rate changes on the appropriate part of the heart rate/R-R interval curve (false negative). This part of the curve is reached during the high intensity stimulation of B+C fibres. Finally and perhaps most importantly, nothing is known concerning the projection patterns of B and C fibres onto their postganglionic targets, the possibility of convergence offers the possibility of occlusion. B and C-fibre preganglionic convergence has been demonstrated in sympathetic ganglia (Janig et al. 1983).

Selective study of unmyelinated fibres has been accomplished with the techniques of: cooling (Ritchie & Straub 1956; Paintal 1965; Patberg et al. 1984), anodal block (Sasen & Zimmerman 1980), heating (Klump & Zimmerman 1980), local anaesthetics (Heavner & De Jong 1974), pressure block (Clark et al. 1935), capsaicin application (Jansco & Such 1983).

Local anaesthetic block does not discriminate adequately between non-myelinated and small myelinated fibres (Nathan & Sears 1961). Differential cooling might offer a simple test of the hypothesis that the bradycardia of the pulmonary chemoreflex is a C-fibre to C-fibre reflex (Daly 1991). It is possible that there is already evidence of this in the literature. In cats the bradycardia induced by PBG has been reported not to be abolished until the vagi are cooled to 3°C (Dawes et al. 1951). It is unknown whether this bradycardia is due to cardiac vagal C-fibre efferents or sympathoinhibition or both. The great advantage of anodal block over differential cooling is that the limitation of C-fibre discharge when A-fibres are blocked is not so marked. This becomes more important if C-fibres have weak chronotropic action. Continuous anodal block or prolonged high frequency stimulation is to be avoided as this leads to nerve deterioration quite rapidly (Whitwam & Kidd 1975;Thoren et al. 1977).

A modified anodal block technique has been described (Accornero et al. 1977) and successfully used to demonstrate the cardioinhibitory actions of vagal C-fibre efferents in the rat (Nosaka et al. 1979) and rabbit (Wooley et al. 1987). The modification in the technique involves the use of triangular shaped pulses to avoid anodal break excitation.

The possibility of asynchronous firing of myelinated axons during the anodal block was eliminated by Accornerro et al. (1977) through the use of the collision test of Douglas and Ritchie (1957).

In cats, it is generally accepted that cardiac slowing is mediated entirely (Middleton et al. 1950; Kidd & McWilliam 1982) or mainly (Heinbecker & Bishop 1935) by myelinated B-fibres and these myelinated axons also seem to mediate most of the reductions in atrial contraction and all of the slowing of A-V conduction evoked by vagal stimulation (Kidd & McWilliam 1982). However, a chronotropic action for C-fibre cardiac vagal efferents has been purported in other species, in the turtle heart (Heinbecker 1931), the rat (Nosaka et al. 1979) rabbit (Heinbecker & Bishop 1935; Wooley et al. 1987) and guinea pig (McWilliam & Wooley 1987). The cardiac division of the fourth branchial nerve of the elasmobranch Scyliorhinus canicula is completely myelinated (Barrett & Taylor 1985b). The comparison of amphibian and fish myelination to that in occurring in the nerves of mammalian species must be made with caution. This is simply because elasmobranchs for instance, inhabit a world of 10 degrees centigrade. This can create the illusion of species differences and effectively mask underlying templates which all vertebrates may share. It is surprising that amongst mammals, there are striking species differences reported, and this concerns such a basic motor system: the vagal control of the heartbeat.

3.1a Pharmacologv of the C-fibre cardiac vagal efferent response in the rabbit

The bradycardia mediated by non-myelinated axons in the rabbit is resistant to the nicotinic ganglion blocker hexaméthonium (Woolley et al. 1987; Ford & McWilliam 1986; McWilliam & Woolley 1990). The bradycardia elicited both by B and C fibres is sensitive to atropine. The slow return to baseline heart rate after C-fibre stimulation and the hexaméthonium resistance suggested to McWilliam et al. (1990), that at the cardiac ganglion a transmitter other than acetylcholine is involved. The inability of nicotinic cholinergic antagonists to abolish ganglionic events is commonly interpreted as evidence of non cholinergic pathways. Seabrook et al. (1990) reported that responses resistant to mecamylamine (cholinergic nicotinic antagonist) in a neonatal rat cardiac ganglion utilize noncholinergic transmission. However it must be emphasised that nicotinic blockers block nicotinic transmission not cholinergic transmission in toto. In fact it turns out that the

sEPSPs in the rat cardiac ganglia recorded by Seabrook et al. (1990) do involve cholinergic receptors but that these are muscarinic not nicotinic. (Selanyko & Skok 1992a,b,c). In fact this nicotinic/muscarinic co-transmission should have been no surprise: many parasympathetic and sympathetic ganglia possess this mechanism. This is the most important and basic aspect of autonomic neurotransmission. Blumberg and Janig (1983) have a described B-fibre sympathetic preganglionic pathways utilizing cholinergic nicotinic transmission to postganglionic sympathetic vasoconstrictor neurones whilst C-fibre preganglionic input is via cholinergic muscarinic transmission. The question arises: are cardiac ganglia organized in a similar fashion? Recent experiments on mammalian cardiac ganglion neurones, cultured or acutely dissociated in vitro, have showed sEPSP synaptic events which are mediated through Mj and M2 receptors. (Allen

& Burnstock 1990; Xi-Moy et al. 1993; Selyanko & Skok 1992b). This phenomenon was first reported by Kuffler’s group in the mudpuppy (Necturus maculosus) (Hartzell et al. 1977). The recent in vitro mammalian work has also revealed interesting facts concerning the organization of the cardiac vagal postganglionic neurones. Xi et al. (1991) have described two populations of principal cardiac ganglion neurones, with different morphologies and electrophysiological properties (these elegant experiments utilized intracellular recording and labelling). One population fires tonically to a step depolarization stimulus, and the other fires in a phasic fashion. It is becoming increasingly clear that most parasympathetic and sympathetic ganglia are organized in a very similar fashion (Saffrey et al. 1992). However, the parallels in the central organization of cranial and sacral parasympathetic preganglionic outflow and the peripheral construction of the ganglia has not featured in the literature;this is perhaps due to the increasing division in the disciplines of in vitro and in vivo electrophysiology.

In 1914, Dale demonstrated that acetylcholine produced two responses that were mimicked by nicotine and by muscarine and that the muscarinic action was blocked by atropine. It is only since 1980 that muscarinic receptor behaviour was better defined through the use of the drug pirenzepine (Hammer & Giachetti 1982). The phrase "muscarinic receptor behaviour" is used, because in 1980 it was unknown whether there were different receptor subtypes or different coupling stratagems used by one receptor and/or different binding sites for antagonists. The techniques of molecular biology have

now confirmed the existence of as many as five different receptor subtypes, and these have been sequenced, cloned and expressed (Hammer & Giachetti 1982). All cardiac ganglion neurones dissociated in cell cultures express muscarinic receptors (Saffrey et al. 1992). In situ hybridization indicates that most of the muscarinic receptor genes are expressed in vitro and in situ. This has been repeatedly demonstrated by autoradiography and electrophysiological analysis (Saffrey et al. 1992).

Receptor sensitivity to the blocking actions of pirenzepine is present in the jpardiac ganglion of the guinea pig (Allen et al. 1990), dog (Xi-Moy et al. 1993) and chicken (Jeck et al. 1988). Block of the nicotinic ganglionic transmission in the chicken heart by itubocurarine unmasked an excitatory muscarinic transmission, which was mediated through Mj-receptors stimulating a low and prolonged release of acetylcholine (Jeck et al. 1988). There is evidence for Mj transmission in airway parasympathetic ganglia (Barnes 1993;Bloom et al. 1988), superior cervical ganglion (Ashe & Yarosh 1984; Brown & Constanti 1980),and enteric nervous system (North et al. 1985).

The neuropeptides; galanin, substance P, CORF, LHRH and NPY have been localized in the cardiac ganglia of a number of species (Knoppe et al. 1992) but best described in the transparent cardiac ganglion of the mudpuppy {Necturus maculosus). No physiological role has been ascribed to any peptide in cardiac ganglia with the possible exception of NPY which mediates vago-sympathetic antagonism (Potter 1987). Peptidergic transmission is characterized by long latency and long duration signalling (Horn 1992; Knoppe 1992). Nicotinic receptors operate on the millisecond timescale (fast EPSP), and muscarinic receptor operation lasts seconds (sEPSP), in a physiological system this means the difference between beat by beat data and breath by breath data. Peptidergic transmission lasts many minutes (late sEPSP), therefore it is unlikely that peptides are of any importance in dynamic cardiovascular control. Also peptidergic transmission cannot account for the lack of respiratory modulation of the pulmonary chemoreflex bradycardia in the cat, since the phenomenon develops rapidly (the first breath) and lasts only for a few seconds before respiratory sinus arrhythmia resumes (personal observation). It may be theorized that the most likely candidate to explain this phenomenon is a muscarinic mechanism, therefore in the following experiments

pirenzepine(an Mj antagonist) was used to test this theoretical prediction.

Although cardiac labelling with HRP delineates DVMN and NA as the source of preganglionic neurones in most species studied (Withington-Wray et al. 1987) there are reported species differences in the chronotropic action of vagal stimulation. It is clear from the literature that there are many problems with the experiments that have yielded this result. It is the object of the present study to investigate the chronotropic action of C-fibre stimulation in anaesthetized rat, cat and rabbit using the modified anodal block technique (Accornero et al. 1977). This technique has been previously applied to the rat (Nosaka et al. 1979) and the rabbit (Woolley et al. 1987) but never to the cat. In addition the hexaméthonium resistance of C-fibre provoked bradycardia in rabbits (Woolley et al. 1987) is tested for pirenzepine resistance. These experiments are conducted to test the hypothesis that C-fibre cardiac vagal preganglionic neurones in the cat, which have already been shown to lack respiratory rhythm (Chapter 2), are involved in the cardioinhibition of the pulmonary chemoreflex.

3.2 Methods

3.2a General

For rats and cats, this is essentially the same as that described in chapter two. Female rabbits (New Zealand Whites) weighing 1.8-2.4kg were anaesthetized with urethane (Sigma)(1.4g/kg) administered via a marginal ear vein or pentobarbitone sodium (Sagatal) i.p. The animals were pretreated with atenolol (1 mg/kg). This selective beta^-antagonist was added to outrule sympatho-inhibition as a contributing mechanism in the evoked bradycardias.

3.2b Preparation of the cats

High cervical bilateral vagotomy was performed and the intact right cranial cardiac branch was placed upon recording electrodes. Stimulating bipolar electrodes were placed upon the cut peripheral end of the right cervical vagus. The inter-electrode distance was 2-3 mm the electrode was made of silver (diam. : 0.5mm).

3.2c Preparation of the rats

The preparation of the rat was identical to the cat. It was possible to locate the cranial

cardiac branch in the rat and this branch when stimulated could arrest the heart and was traced to cardiac ganglia in the fat pads surrounding the heart. The cardiac branch was found to be readily accessible without resort to lobectomy. The fat pads were dissected in vitro after the experiment. The ganglia were made visible by applying 1 % neutral red. The ganglia were stored in 10% formal saline and then paraffin sectioned 3 days later. The sections were stained for Nissl substance.

3.2d Preparation of the rabbits

It was difficult to maintain rabbits in an acceptable state post thoracotomy. This approach was abandoned, and the recording electrodes were placed not on the cardiac branch but on the lower cervical vagus. In other respects the preparation of the rabbits was the same as that for the cats and rats.

3.3 Results

The modified anodal block technique of Acornerro et al. (1977) was found to be a simple, convenient method for selectively stimulating C-fibres. Care was taken to ensure good quality unambiguous action potential recordings (fig 3.1, fig 3.2). Signal averaging techniques were used to accentuate possible myelinated break-through and subtract movement artefact from the cardiac branch records. This was considered insufficient however, since the main concern is for a single or multiple asynchronous volley breakthrough which the signal averaging process will actually mask. This is an important issue since B-fibre preganglionic efferents are extremely potent even in small numbers. Therefore every sweep was individually scrutinized (200 sweeps per stimulation protocol i.e. lOHz for 20 seconds). The results dramatically illustrate the species similarities (fig 3.3) (fig 3.5). Greater certainty of the potency of the anodal block can be obtained through single volley stimuli. The experiments of Brown & Eccles (1934) were revisited (fig 3.6 & 3.7). The state of modern computing means that hundreds of thousands of cardiac cycle lengths can be calculated in seconds, a feat that would have taken the early experimenters months to perform manually.

FIGURE 3.1

Experimental arrangement for anodal block technique. Atenolol (1 mg/kg) treated animals had bilateral vagotomy performed. Stimulating electrodes were placed on the cut end of the vagus. In order to stimulate all the fibres the cathode faced the heart, the evoked compound action potential (this is from a rabbit) displays A,B and C waves. When the polarity is switched and a triangular shaped pulse is applied, the A and B waves are completely blocked and the amplitude of the C wave is partially reduced. The dot indicates the point of stimulation.

T